Medical Image Captioning on Lung Cytology (Malignant) benchmark leaderboard · SOTA2 Research

Medical Image Captioning on Lung Cytology (Malignant)

0.853BLEU-1Vision Model + Dual Text Decoders

Evaluation Results

Method
Vision Model + Dual Text Decoders2024.03	0.853	0.804	0.756	0.724	0.494	0.844	4.883	0.718
BLIP22024.03	0.844	0.783	0.721	0.671	0.469	0.833	2.879	0.739
CNN + single text decoder2024.03	0.835	0.777	0.722	0.685	0.471	0.816	4.37	0.689
BLIP2024.03	0.835	0.777	0.719	0.672	0.467	0.828	3.01	0.717
BLIP2024.03	0.831	0.766	0.701	0.653	0.457	0.812	2.939	0.695
BLIP22024.03	0.824	0.753	0.683	0.631	0.448	0.807	2.791	0.688
GIT2024.03	0.821	0.752	0.685	0.636	0.448	0.803	3.073	0.691
GIT2024.03	0.81	0.745	0.681	0.633	0.446	0.797	3.139	0.676